Somebody has to write it down. Then somebody has to amend it.

Every new league has the same conversation in August. Somebody proposes a scoring setting, somebody else agrees, a third person misremembers the exchange by October. By December, when a playoff seeding hangs on a tight end's touchdown catch is worth the same as a running back's touchdown run, the founding mythology of the league has already split into two incompatible versions. The constitution exists to make that argument unnecessary — not to prevent disagreement, but to make it resolvable by reference to a text rather than memory.

A league constitution is simply a written record of what a league has agreed to, with a procedure for changing it. That second part is the part most leagues skip, and it is the part that matters most. A document with no amendment procedure is not a constitution; it is a transcript. It records what was decided and then becomes a trap, because the rules that seemed obvious in Year One will not survive contact with a running back whose team changes its offensive scheme, a scoring platform that adds a new statistical category, or a keeper price that turns a third-round pick into the best contract in the league by November.

The NFL's own collective bargaining agreement — the contract between the league and its players — runs to hundreds of pages and is renegotiated on a fixed schedule, with formal procedures for what happens when the parties cannot agree. That is not coincidence. Complex systems with distributed stakes need written rules and written rules need a mechanism for revision. A fantasy league is not the NFL, but the structure of the problem is identical: multiple parties, real stakes, rules that interact in ways nobody fully anticipated when they were written.

The core of any keeper league constitution is the roster: how many players, in what positions, on what timeline. Alongside that sits the draft format — auction or snake, how many rounds, when in the offseason — and the scoring settings in enough detail to resolve edge cases. These sections feel easy to write because they mostly describe things the league has already decided. The hard sections are the ones describing things that have not happened yet.

Keeper rules belong in the hard category. A league that keeps players by surrendering the round they were drafted in needs to specify what happens to an undrafted free agent signed mid-season: what round does he carry? A league using an escalating salary needs to specify what salary a player enters at if he was a waiver-wire acquisition rather than a draft pick. A league using contract terms needs to specify what happens to a contract if the player retires, is cut by his NFL team, or misses the season to injury. These are not edge cases; they are things that happen every year in a thirty-two-team league with an eighteen-week season and a running back position whose average career length runs under three years. The constitution that does not answer them will be asked to answer them at the worst possible moment.

The amendment procedure itself should specify three things: who may propose a change, what threshold is required to pass it, and when in the league year a change may take effect. That last clause is the one leagues forget and then regret. A rule change ratified in Week 11 of a competitive season is not a rule change; it is a lever someone pulled to advantage themselves, or at minimum it will be perceived that way. The standard practice among long-running leagues is to allow amendment only in the offseason — after the championship, before the draft — so that no change affects the stakes of a season already in progress. That window is also when the effects of the current rules are freshest and the disagreements most productive.

The commissioner's powers need their own section. Not because commissioners are untrustworthy, but because undefined power produces resentment in exact proportion to how often it is exercised. A commissioner who makes a judgment call that saves a playoff bracket will be thanked once and then resented the next time he makes a different call. A constitution that specifies what decisions the commissioner may make unilaterally, which require a vote, and what the vote threshold is, turns a personal judgment into an institutional process. The league runs the same rule on the commissioner that it runs on everyone else.

The leagues that break apart usually do so not over a single catastrophic disagreement but over accumulated ambiguity. A rule that seemed clear becomes contested; the commissioner rules one way; a faction disagrees; the faction leaves. What is lost is not just the departing members but the institutional knowledge they carried — the memory of why the rule was written the way it was, who it was designed to protect, what the alternative was.

The 2011 NFL lockout exposed this problem at scale for fantasy leagues. Leagues whose constitutions assumed an uninterrupted preseason calendar discovered their own rules did not specify what happened to keeper deadlines, contract years, or salary escalators when the offseason ceased to exist. The leagues that navigated it cleanly were the ones that had either written contingency language in advance or had an amendment procedure robust enough to pass emergency rules before the regular season began. The leagues that argued all summer were the ones that had neither.

A constitution's amendment procedure is most valuable precisely when the league least wants to use it: when the season is close, the stakes are real, and two reasonable people read the same sentence and come away with different meanings. The procedure does not make the disagreement go away. It makes the disagreement resolvable in a way that all parties agreed to in advance, which is the closest a fantasy league gets to jurisprudence.

Write the amendment procedure in August, when nobody needs it. Because the version of you sitting in a Week 13 playoff argument, staring at a rule that does not cover this situation, would give a lot for a document that tells you what to do next — and more for one that tells you who gets to decide.